Improving the photovoltaic parameters in CdS quantum dot sensitized SnO 2 based solar cells through incorporation of chemically deposited compact SnO 2 layer

: Mitigation of the recombination of electrons within the quantum dot sensitized solar cells (QDSSCs) and hence the improvement in the performances of the devices can be achieved by incorporation of a compact layer in between the transparent conducting substrate and the semiconducting materials used in these devices. In this work, a facile and cost-effective method of incorporation of a compact layer of SnO 2 over the Fluorine doped Tin oxide (FTO) substrate and its effect on the efficiency enhancement of the CdS sensitized SnO 2 QDSSCs have been studied by means of current-voltage characteristics, Electrochemical Impedance Spectroscopy, and electron lifetime estimation. The incorporation of SnO 2 compact layer improved the overall efficiency of the device by 250% as compared with the devices fabricated with no compact layer under the illumination of 100 mW cm -2 . The improvement in the open-circuit voltage and the significant enhancement in the short circuit current density (~200%) together with the increase in the electron lifetime in the QDSSC with compact layer suggested that the compact layer has acted as a weak energy barrier, which increased the electron density in the mesoporous SnO 2 film. The enhancement in the short circuit current density and the efficiency mainly stems due to the decrease in the series resistance and the increase in the recombination resistance of the device fabricated with the compact layer.


INTRODUCTION
Within the third-generation photovoltaic cells, quantum dots (QDs) based solar cells (QDSSCs) have recently attracted much attention due to the distinct properties of the quantum dots. High molar extinction coefficients, ability of multiple exciton generation, high stability to heat, tunable energy gap due to the quantum confinement effect, and high theoretical efficiency are some of the major advantages of quantum dots based solar cells (Tian and Cao, 2013;Pan et al., 2018;Zahu et al., 2020). Quantum dots such as CdS (Chang and Lee, 2007;Chen et al., 2009;Sudhagar et al., 2009 ;Tak et al., 2009;Zhu et al., 2011), CdSe (Robel et al., 2007;Chen et al., 2010) and PbS (Hoyer and Konenkamp, 1995;Plass et al., 2002;Leventis et al., 2010) are widely used in these QDSSCs. Among these QDs, CdS has attracted a wide interest due to its suitable optical properties for the solar cells, such as efficient light photon absorption in the visible region of the solar spectrum (Nozik, 2002;Chang and Lee, 2007;Chen et al., 2009;Zhu et al., 2011). Typically, in the construction of QDSSCs, suitable quantum dots have been used to sensitize a semiconducting material such as TiO 2 . Improvement in the power conversion efficiencies (PCE) of QDSSCs from less than 5% to 12% over have been reported with the usage of complex quantum structures and modifications of photoanode over the past decade (Du et al., 2016;Zhang et al., 2019;Pan et al., 2019;He et al., 2020). However, the efficiencies of CdS sensitized solar cells fabricated without any post treatments, combinations or modifications on QDs are still lying in the range of 0.7-1.1% (Kim et al., 2011;Zhou et al., 2013). Since these QDs do not interconnect via chemical bonding with the semiconducting material, the structure and the morphology of the photoanode are a crucial factor determining the power converting efficiency (PCE) of the device especially in the charge transfer processes (Tian and Cao, 2013;Pan et al., 2018;Sahu et al., 2020;). Photoanodes fabricated with TiO 2 based substrates are currently the most attractive structures in fabrication of QDSSCs. This is due to the higher efficiencies achieved compared to other semiconductors such as SnO 2 or ZnO (Ju et al., 2010;Tian and Cao, 2013). However, further improvements in the device performances have been limited due to the low electron mobility and transport properties of the TiO 2 (Hendry et al., 2006). In this scenario, SnO 2 is a promising candidate having a wide band gap (3.6 eV) and higher electronic mobility. Moreover, its low sensitivity to UV degradation and better long term stability, make the SnO 2 as a very promising material for these devices (Park et al., 2004). However, application of SnO 2 as a photo anode is confined due to the poor adhesion to the substrate materials like Fluorine doped Tin oxide glasses (FTO). Moreover, comparatively little work has been carried out on SnO 2 to achieve the results suggested by theoretical studies. Most of the devices showing comparable efficiencies with the TiO 2 photoanodes are fabricated with a combination of CdS with CdSe (Liu et al., 2020). However, to the best of our knowledge, only very few studies have been carried out with CdS sensitized solar cells with SnO 2 . In this context, Zhou et al. (2013) have reported CdS sensitized SnO 2 solar cells with 0.22% PCE and very recently Liu et al. (2019) have reported CdS sensitized sol-gel derived thin films of self-patterned micro-blocks of closely-packed SnO 2 nanoparticles as high-performance photoanodes in alkaline solution of methanol with higher photo current density. However, overall device performances with PCE were not reported. In this study, a novel photoanode structure comprising with compact layer of SnO 2 was introduced in the first time in the fabrication of CdS sensitized SnO 2 solar cells towards the efficiency enhancement. The device performances have been compared with and without the compact layer in the photoanode.

Fabrication of SnO 2 electrode
SnCl 2 (2.0 g) was ground well with 4.0 mL of 1 M H 2 SO 4 for 15 minutes. SnCl 2 reacted with H 2 SO 4 to form SnSO 4 . This paste was spin coated on the pre cleaned FTO at 3000 rpm for 60 seconds. Then, the SnSO 4 coated FTO glass was sintered at 500 ºC for 2 hours. At 500 ºC, SnSO 4 oxidized and SnO 2 was formed. For future reference, this layer is referred as the SnO 2 compact layer (CL). Then, 3.0 mL of SnO 2 colloidal suspension, 10 drops of glacial acetic acid, and 0.02 g of MgO were ground well. Then, 5 drops of Triton X-100 and 40.0 mL of absolute ethanol were added to the mixture. The resulting nanoparticle (NP) colloidal sample was sonicated for 20 minutes. FTO/SnO 2 CL was placed on a hot plate at 150 ºC. Then the colloidal mixture was sprayed over for several times. Then, the SnO 2 CL/ SnO 2 NP electrode films were sintered at 500 ºC for 30 minutes.

Deposition of CdS quantum dots
CdS quantum dots were deposited on the SnO 2 electrode using successive ionic layer adsorption and reaction (SILAR) method. 0.4 M cadmium chloride (CdCl 2 ) and 0.1 M sodium sulfide (Na 2 S) aqueous solutions were prepared for cationic (Cd 2+ ) and anionic (S 2-) precursors respectively. For the SILAR method, the SnO 2 film was dipped in the cationic precursor for 1 minute and thoroughly washed in ethanol : deionized water (1:1) mixture. The same procedure was carried out with the anionic precursor with the same dipping time. 10 SILAR cycles were carried out to fabricate the best CdS quantum -dot sensitized photoanode.

Preparation of the polysulfide electrolyte
Liquid polysulfide electrolyte was prepared by the following method. 2 M sulfur and 2 M Na 2 S were dissolved in a mixture of deionized water and methanol in the ratio of 3:7 (v/v). The mixture was stirred for 30 minutes.

Solar Cell assembly
An appropriate amount of the polysulfide electrolyte was applied on to the CdS sensitized photoanode and then sandwiched it with the platinized counter electrode as schematically shown in Figure 1. Platinum coated FTO glass substrate was used as the counter electrode. Then the solar cell was clipped together using couple of steel clips.

Current -Voltage characterization
In order to compare the photovoltaic performance of the QDSSCs, with and without the compact layer of SnO 2 , current density-voltage (J-V) measurements of solar cells were measured under the illumination of 100 mW cm -2 with AM 1.5 spectral filter using a computer controlled multimeter (Keithley 2000 model) coupled with potentiostat/ galvanostat unit (HA-301). The active area of the QDSSC was 0.25 cm 2 .

Electrochemical Impedance Spectra (EIS) measurements
In order to determine the interfacial resistance values of the solar cells, electrochemical impedance spectra of the QDSSC was obtained by using PGSTAT128 N with an FRA 32M Frequency Response Analyzer (Metrohm) under the illumination of 100 mW cm -2 (Solar simulator with AM 1.5 spectral filter) in the frequency interval from 0.01 Hz to 1 MHz. Carrier transport resistance, recombination resistance and series resistance of the interfaces of the QDSSCs were calculated by fitting the electrochemical impedance spectroscopy (EIS) data with an appropriate equivalent circuit of the QDSSC. Electron lifetimes in the solar cells were estimated with Bode phase plots.  with no compact layer in the photoanode, while curve (2) shows the same characteristics with the photoanode with SnO 2 compact layer under the illumination of 100 mW cm -2 . Device parameters namely, short circuit current density (J SC ), open circuit voltage (V OC ), fill factor (FF) and the power conversion efficiency (η) determined from those curves and are tabulated in Table 1. The highest efficiency of 0.77% is achieved from the QDDSC fabricated with the photoanode, comprised with a compact SnO 2 layer which is ~ 250% higher than that of the device assembled with no compact layer in the photoanode with the same thickness of mesoporous SnO 2 layer. As it is evident from these two curves and also from the values tabulated in the Table 1, both the V OC and J SC of the CdS QDSSCs have been greatly enhanced upon the addition of the SnO 2 compact layer. However, the most noticeable is the enhancement in the J SC , which has increased from 2.28 mA cm -2 to 6.82 mA cm -2 . The fill factor has not been drastically affected by the novel SnO 2 layer. This improvement in photovoltaic performance of compact layer based DSSC could be attributed to the decrees in the electron recombination within the device. The efficiency of the reference cell with no compact layer was recorded to be 0.21%. This is in well agreement with the efficiency value of 0.22% reported by Zhou et al. (2013) in CdS sensitized SnO 2 solar cells.

Electrochemical impedance spectroscopy (EIS)
In order to get an idea about the electronic and ionic processes and recombination resistances of the QDSSCs, Electrochemical impedance spectroscopy (EIS) measurements were performed on the devices fabricated with and without the compact layer of SnO 2 . Figure 3 shows the Nyquist plots of the CdS sensitized QDSSCs and the equivalent circuit model used. As it is depicted in the figure, two characteristic semicircles including a smaller one at higher frequency and a larger one at intermediate frequency can be seen. In equivalent circuit, R S is the series resistance which represents the FTO/SnO 2 interface, while R 1CT is the resistance at the counter electrode/electrolyte interface and R 2CT is the resistance at the photoanode/electrolyte interface which is known as the recombination resistance. CPE 1 and CPE 2 are the constant phase elements and W is the finite Warburg impedance element which is related with diffusion process. Low recombination resistance favors the electron recombination, which negatively affects the cell performance. Hence, a higher recombination resistance is favorable to higher cell performance. EIS parameters are listed in the Table 2  . Table 2 shows the impedance parameters of the devices estimated from the above equivalent circuit. As it is evident from the Table 2, the series resistance R S of the QDSSC fabricated with SnO 2 compact layer in the photoanode is lower than those of the device fabricated without such a compact layer in the photoanode. On the other hand, significant increase in both charge transfer resistances R 1CT and the R 2CT can be observed. Even though there is no evidence to correlates the increase in the R 1CT with the enhancement of cell performances, the increase in the R 2CT , which represents the recombination resistance can be correlated with the increase in the J SC as well as in the V OC as tabulated in the Table 1. Therefore, EIS data are in correlation with the increase in the efficiency due to the enhancement in both J SC and V OC of the QDSSC with SnO 2 compact layer in the photoanode (Dissanayake et al., 2020).    Figure 4 shows the Bode phase plots obtained based on the EIS measurements of CdS QDSSCs with and without SnO 2 compact layer. Life-time of the electron τ can be calculated from the frequency corresponding to the maximum phase angle f max using the following equation  and estimated values are tabulated in Table 3 The electron lifetime is directly proportional to the recombination resistance (R 2CT ), which is consistent with the data. Further, the higher electron lifetime positively affects the photocurrent as well . As shown in the Table estimated electron life-time of the device fabricated with the compact layer is higher than that of the device without a compact layer in the photoanode. This means that in this novel photoanode configuration electrons diffuse and transfer more easily than that of the photoanode with no compact layer. Higher electron lifetime also implies reduced recombination of photoelectrons between SnO 2 photoanode and the electrolyte, which contributes to the enhancement in J SC leading to overall efficiency enhancement .     In order to show the above charge transfer process within the device, band positions and possible charge transfer mechanism are schematically shown in Figure 5 (Lee and Chang, 2008;Kim et al., 2011;Zhou et al., 2013).
As it is well known when the QDSSCs are exposed to the sunlight, the QDs harvest some photon energy in the light corresponding to their band gap values. Accordingly in this case CdS QDs with 2.3 eV band gap absorb the corresponding photons and generate the photo-excited electrons. These excited electrons are then injected from its conduction band (CB) to the CB of the SnO 2 semiconductor and then subsequently migrated to the external circuit through the conductive substrate where SnO 2 is deposited (FTO). These electrons then migrate towards the counter electrode. Meantime holes produced by the photoexcitation process synchronously in the valence band (VB) of the QDs, are immediately transferred to the redox electrolyte to oxidize it. The oxidized electrolyte obtained the electrons in the counted electrode from external circuit in the device. One of the crucial factors determining the overall efficiency is the electron separation and transport before recombining within the device. Therefore, in the present case due the existence of the compact SnO 2 layer in between the FTO and the porous SnO 2 the back electron transfer from the FTO substrate must be minimized as confirmed by the EIS measurements.

CONCLUSION
In summary, here we report the introduction of the compact SnO 2 layer by a simple chemical solution deposition method, which is instrumental in improving the overall photovoltaic performance of the SnO 2 based CdS quantum dot sensitized solar cells. Incorporation of this compact layer resulted in 250% efficiency enhancement in CdS sensitized SnO 2 based solar cells favoring the forward direction electron transfer and prevents the electron recombination either or both with the redox mediator and the holes produced in the CdS quantum dots.